AI Benchmark Leaderboard

Compare AI model scores, coding ability and performance in one place.

# Model Org Intelligence ↓ CodingMathMMLU ProGPQALiveCodeBenchAIME 2025MATH 500SciCodeIFBenchHLE
126 Nemotron 3.5 Super VL NVIDIA 13.6 ——————————
127 Inkling Small Thinking Machines 12.9 ——————————
128 DeepSeek V4 Pro 0813 (Non-reasoning) DeepSeek 12.6 ——————————
129 Inkling (Xhigh) Thinking Machines 12.5 ——————————
130 Ministral 3B (Dec '25) Mistral 12.1 4.8 22.0 52.4 35.8 24.7 22.0 — 14.4 26.8 5.3
131 Nemotron 3 Ultra 550B A55B (Reasoning) NVIDIA 11.8 ——————————
132 Gemini 3.5 Flash-Lite Google 11.7 ——————————
133 Qwen3.8 27B (Non-reasoning) Alibaba 10.0 ——————————
134 GPT-6 Luna (Low) OpenAI 9.7 ——————————
135 GPT-6 Luna (Non-reasoning) OpenAI 9.4 ——————————
136 K-EXAONE 2.0 LG AI Research 7.6 ——————————
137 K-EXAONE 2.0 LG AI Research 7.6 ——————————
138 Claude 4.5 Haiku (Reasoning) Anthropic 7.4 ——————————
139 Muse Glimmer (High) Meta 6.3 ——————————
140 Command A+ Cohere 6.2 ——————————
141 Gemini 3.1 Flash-Lite Google 6.2 ——————————
142 Nemotron 3.5 Lightning NVIDIA 5.5 ——————————
143 Granite 4.2 8B IBM 5.1 ——————————
144 Granite 4.2 3B IBM 4.7 ——————————
145 Mistral Medium 3.5 Mistral 4.6 ——————————
146 Gemma 4 31B (Reasoning) Google 4.4 ——————————
147 Mistral Small 4 (Reasoning) Mistral 4.3 ——————————
148 gpt-oss-120b (High) OpenAI 3.8 ——————————
149 North Mini Code Cohere 3.4 ——————————
150 K2 Horizon 0.9B MBZUAI Institute of Foundation Models 3.0 3.4 —— 29.3 ——— 6.9 — 5.4

AI Benchmarks FAQ

Which AI model is best for my task?

Compare the scores most relevant to your task, then consider speed, pricing and practical requirements. A high score in one benchmark does not make a model the best choice for every use.

What does a missing benchmark score mean?

A dash means that a score is not available for that model in this table. It should not be treated as a score of zero.

How are the top models ranked?

By the intelligence index, highest first. The Top models cards show the eight highest-scoring current models, one per model family, and models with the same score share a place.

Where do the scores come from?

From independent third-party model evaluations, which Franklin AI syncs automatically each day. Each model page links to About these benchmarks, which explains them in more detail.

What does a model page show?

Its scores, its price per million input and output tokens, its measured speed and comparisons with other models, where that data is available.